xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 02:25:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Doc] Remove Useless Comments (#24687)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye 2025-09-11 15:25:47 -04:00 committed by GitHub
parent 127ded0a9e
commit a892b259b4
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
vllm/model_executor/layers/quantization/fp8.py
View File

@ -757,10 +757,9 @@ class Fp8MoEMethod(FusedMoEMethodBase):
layer.w2_weight = torch.nn.Parameter(shuffled_w2,
requires_grad=False)
# DeepGemm scales need to be transposed and aligned. We try to do
# DeepGemm scales need to be transposed and aligned. We try to do
# it ahead of time for performance reasons.
if self.allow_deep_gemm and not is_deep_gemm_e8m0_used():
# Lazy import to avoid CUDA initialization problems.
if _is_col_major(layer.w13_weight_scale_inv):
layer.w13_weight_scale_inv = \
get_col_major_tma_aligned_tensor(layer.w13_weight_scale_inv).contiguous()